Is antibiotic resistance reversible? Evidence from humans and food animals Anna Mae Scott, Mina Bakhit, Chris Del Mar Centre for Research in Evidence-Based Practice (CREBP) Bond University, Australia
Antibiotic resistance: a problem. Antibiotic resistance = direct consequence of antibiotic use Consequences of antibiotic-resistant infections: Currently: est. 700,000 lives/year globally By 2050: est. 10M lives/year globally O'Neill, J., S. Davies, and J. Rex, Antimicrobial Resistance: Tackling a crisis for the health and wealth of nations 2014.
Source of the problem 1) Pace of discovery of new antibiotics has slowed 2) Antibiotic use is rising - In humans - In animals Jim O Neill et al, 2015, Antimicrobials in Agriculture and The environment: Reducing unnecessary Use and waste
3 questions to ask 1) Does decreasing antibiotic use in animals, decrease resistance in animals? 2) Does decreasing antibiotic use in animals, decrease resistance in humans? 3) Does decreasing antibiotic use in humans, decrease resistance in humans?
Questions 1 & 2: WHO systematic review Systematic review of evidence for WHO in 2016 Aim: To underpin the guideline to preserve the long-term effectiveness of antimicrobials critical for human medicine Focus: Does limiting the use of antimicrobials in food animals reduce resistant elements in 1) food animals; and 2) humans
Questions 1 & 2: WHO systematic review No date, language restrictions Study designs included: Animal question: cohorts, RCTs, reviews Human question: before & after, ITS, cohorts, RCTs, reviews
3 questions to ask 1) Does decreasing antibiotic use in animals, decrease resistance in animals? 2) Does decreasing antibiotic use in animals, decrease resistance in humans? 3) Does decreasing antibiotic use in humans, decrease resistance in humans?
Question 1: WHO systematic review 89 studies included Adequate evidence to conclude that limiting ABs in animals reduces resistance in animals N.B. considerable heterogeneity: design, species, isolates, environs, ABs, routes of administration, sampling timeframes, methods precludes estimating the magnitude of effect
3 questions to ask 1) Does decreasing antibiotic use in animals, decrease resistance in animals? 2) Does decreasing antibiotic use in animals, decrease resistance in humans? 3) Does decreasing antibiotic use in humans, decrease resistance in humans?
Question 2: WHO systematic review 4 studies included 1 study directly answering the question: Dutil 2010 Interrupted time series Quebec, Canada Ceftiofur injections into eggs in chicken hatcheries Voluntary withdrawal (2005) Subsequent partial reintroduction (2007) Dutil, et al. 2010, Ceftiofur Resistance in Salmonella enterica Serovar Heidelberg from Chicken Meat and Humans, Canada
Question 2: WHO systematic review Dutil, et al. 2010, Ceftiofur Resistance in Salmonella enterica Serovar Heidelberg from Chicken Meat and Humans, Canada
Question 2: WHO systematic review
3 questions to ask 1) Does decreasing antibiotic use in animals, decrease resistance in animals? 2) Does decreasing antibiotic use in animals, decrease resistance in humans? 3) Does decreasing antibiotic use in humans, decrease resistance in humans?
Question 3: Costelloe systematic review Focus: impact of AB prescribing in primary care, on AB resistance in patients Findings: Exposure to antibiotics causes resistance The resistance is the greatest right after exposure but resistance decreases over time Costelloe et al, 2010, Effect of antibiotic prescribing in primary care on antimicrobial resistance in individual patients: systematic review and metaanalysis
Question 3: Costelloe systematic review
Question 3: Costelloe systematic review Costelloe review now 7 years old Literature searches nearly 8 years old Moreover: Cannot replicate the search strategy or the results Re-doing it anew
Conclusions: Is antibiotic resistance reversible? Limiting the use of antibiotics in humans and animals decreases resistance Usefulness of antibiotics can be conserved by reducing their use Caveats: evidence base somewhat limited evidence quality is variable cannot quantify the effect Role for surveillance and reporting systems